The present invention relates to methods for use in the identification of molecules which bind to a target component of a biochemical system or modulate the activity of a target.
Over the past decade, high-throughput screening (HTS) of compound libraries has become a cornerstone technology of pharmaceutical research. Investment into HTS is substantial. A current estimate is that biological screening and preclinical pharmacological testing alone account for ˜14% of the total research and development (R&D) expenditures of the pharmaceutical industry (Handen, Summer 2002). HTS has seen significant improvements in recent years, driven by a need to reduce operating costs and increase the number of compounds and targets that can be screened. Conventional 96-well plates have now largely been replaced by 384-well, 1536-well and even 3456-well formats. This, combined with commercially available plate-handling robotics allows the screening of 100,000 assays per day, or more, and significantly cuts costs per assay due to the miniaturisation of the assays.
HTS is complemented by several other developments. Combinatorial chemistry is a potent technology for creating large numbers of structurally related compounds for HTS. Currently, combinatorial synthesis mostly involves spatially resolved parallel synthesis. The number of compounds that can be synthesised is limited to hundreds or thousands but the compounds can be synthesised on a scale of milligrams or tens of milligrams, enabling fill characterisation and even purification. Larger libraries can be synthesised using split synthesis on beads to generate one-bead-one compound libraries. This method is much less widely adopted due to a series of limitations including: the need for solid phase synthesis; difficulties characterising the final products (due to the shear numbers and small scale); the small amounts of compound on a bead being only sufficient for one or a few assays; the difficulty in identifying the structure of a hit compound, which often relies on tagging or encoding methods and complicates both synthesis and analysis. Despite this split synthesis and single bead analysis still has promise. Recently there have been significant developments in miniaturised screening and single bead analysis. For example, printing techniques allow protein-binding assays to be performed on a slide containing 10,800 compound spots, each of 1 nl volume (Hergenrother et al., 2000). Combichem has so far, however, generated only a limited number of lead compounds. As of April 2000, only 10 compounds with a combinatorial chemistry history had entered clinical development and all but three of these are (oligo)nucleotides or peptides (Adang and Hermkens, 2001). Indeed, despite enormous investments in both HTS and combinatorial chemistry during the past decade the number of new drugs introduced per year has remained constant at best.
Dynamic combinatorial chemistry (DCC) can also be used to create dynamic combinatorial libraries (DCLs) from a set of reversibly interchanging components, however the sizes of libraries created and screened to date are still fairly limited (≦40,000) (Ramstrom and Lehn, 2002).
Virtual screening (VS) (Lyne, 2002), in which large compound bases are searched using computational approaches to identify a subset of candidate molecules for testing may also be very useful when integrated with HTS. However, there are to date few studies that directly compare the performance of VS and HTS, and further validation is required.
Despite all these developments, current screening throughput is still far from adequate. Recent estimates of the number of individual genes in the human genome (˜30,000) and the number of unique chemical structures theoretically attainable using existing chemistries suggests that an enormous number of assays would be required to completely map the structure-activity space for all potential therapeutic targets (Burbaum, 1998).
Hence, the provision of a method which permits screening vast numbers (≧1010) of compounds quickly, using reaction volumes of only a few femtoliters, and at very low cost would be of enormous utility in the generation of novel drug leads.
Tawfik and Griffiths (1998), and International patent application PCT/GB98/01889, describe a system for in vitro evolution using compartmentalisation in microcapsules to link genotype and phenotype at the molecular level. In Tawfik and Griffiths (1998), and in several embodiments of International patent application PCT/GB98/01889, the desired activity of a gene product results in a modification of the genetic element which encoded it (and is present in the same microcapsule). The modified genetic element can then be selected in a subsequent step.